Although drug use and HIV rates among older drug users are increasing at alarming rates, knowledge on drug use patterns and risk behaviors among this cohort as older adults is limited. National trend data show that adults who are 45 or older comprise the fastest growing age group of drug users as well as new drug treatment admissions and new HIV/AIDS cases. The two-fold goal of this exploratory R21 application is to address the lack of current knowledge on older drug users and provide innovative mathematical models that have the potential to inform future research on specific turning points in drug use and related health risks over the life course. In this mixed-methods study we use quantitative methods to identify turning points in drug use and risk behaviors, qualitative methods to explore these turning points more thoroughly, and mathematical analysis to build predictive models on turning points in drug trajectories. The community-based sample will consist of active users (n=50) and former users (n=50) of heroin, cocaine, and methamphetamine age 45 and older. Data are collected using a quantitative matrix on drug history and social roles and a qualitative in-depth interview. The study site is the metropolitan statistical area (MSA) of Atlanta that includes urban, suburban and rural settings. The knowledge gained from an in-depth exploration of this sample will add to our understanding of an aging drug-using population, inform future research on older drug users and provide innovative mathematical models on risk behaviors and drug use to be tested in future large-sample studies. To accomplish this we propose three specific aims: (1) To identify turning points in the onset, continuation, and cessation of drug use throughout the life course of a sample of older users;specifically how social roles, race, gender, age, social contexts, policies and historical events influence turning points in drug use and drug- related HIV risk behaviors;(2) To thoroughly explore these turning points and transitions over the life course, specifically changes in drug availability, risk behaviors, routes of administration, social roles, networks, support, policies, settings, and geographic locations, imbedded in a life course perspective;(3) To build Dynamic Bayesian Networks that best model the static and dynamic aspects found in our quantitative and qualitative data. In so doing we are able to identify not only statistical relationships between the variables but also the influence and progression of these through time. Together these aims provide needed in-depth details on the lives of older drug users as well as trajectory models that can be used to target treatment strategies on the turning points in drug use trajectories. The information and models provided by this study can be used to develop better prevention, intervention and treatment programs by focusing on specific turning points in drug careers. This is a unique and innovative exploration of methodologies that can lead to future studies. The models developed from this R21 exploratory research grant will be used to propose a R01 grant to extend the predictive value of the models on turning points in drug trajectories throughout the life course. PUBLIC HEALTH RELEVANCE: Recent health studies show an alarming increase in drug use and HIV/AIDS among older adults, a cohort composed primarily of aging baby boomers. Retrospective life histories of older adult drug users allow an in- depth exploration how risk factors are influenced by the social context of drug users over the life course and help to identify turning points in the drug trajectory. By applying what we learn from the lives of older drug users in this sample, the findings will inform research and health initiatives for older as well as younger users, including prevention, intervention and treatment programs.